Biology Reference
In-Depth Information
Genomic Selection
major crops. The importance of the genotype
x environment interaction in yield and sucrose-
related traits may explain the slower progress
observed in sugarcane breeding as well as the
highly quantitative nature of trait determinism.
Cultivar improvement has relied largely on tradi-
tional breeding methods. These methods involve
a lengthy (12-15 years) and expensive process
of selection of plants with the desired agronomic
traits. Molecular genomics should help improve
programs by generating molecular markers that
can assist the breeding process or the introgres-
sion of new genes into breeding germplasm.
However, the development of genomic applica-
tions in sugarcane breeding programs has lagged
behind many of the diploid plant models. Adapt-
ing genomic approaches developed for model
plants to sugarcane is rarely straightforward
owing to the complication of its high ploidy level.
In the last two decades many QTL studies based
on bi-parental crosses revealed many QTLs with
small effects, while validation of the stability of
a QTL's effect across genetic backgrounds, time,
and environments remains a challenging task.
Conventional MAS approaches are useful only
for tagging alleles of major effects related to dis-
ease resistance (Al Janabi et al. 2007; Costet et al.
2012), but not for complex traits such as cane
yield or quality characters. Moreover traditional
marker systems are not powerful or convenient
enough to densely scan the sugarcane genome
and to develop efficient routine applications.
Given the prohibitive cost of conventional
programs, the entire sugarcane research com-
munity has a major interest in multidisciplinary
studies aimed at unraveling the determinism
of complex traits in order to design molecular
breeding approaches tailored to 'real life' breed-
ing. The main expectation is more advanced
incorporation of association genetics-based
approaches in sugarcane improvement. In
the context of sugarcane polyploidy, specific
scientific creativity will be needed to design a
breeder-friendly marker system as well as appro-
priate bioinformatics pipes adapted to routine
genotyping. A major innovation in sugarcane
Recently a new MAS strategy, Genomic Selec-
tion (GS), became popular in animal breed-
ing, after the seminal paper of Meuvissen
and colleagues (2001). This strategy simulta-
neously estimates all marker effects across the
entire genome. In this case, instead of many
conventional marker-assisted selection (MAS)
approaches, there is no defined subset of signifi-
cant markers. Marker effects are estimated from
a training panel using phenotypic and genotypic
panel data and pedigree or kinship information.
The accuracy of phenotype prediction is tested
on a validation panel. The concept and applica-
tion of GS need to be investigated in plant breed-
ing (Heffner et al. 2009). To date, a few studies
have been published on plants, most of which
explore the potential benefits of GS from a the-
oretical viewpoint, based on simulations that are
not tailored to any particular biological breeding
context, except for a few studies (Bernardo and
Yu 2007; Wong and Bernardo 2008). GS's poten-
tial resides in the use of all markers as predictors
of complex trait performance, thereby capturing
more of the variation due to QTLs that are hardly
significant. The training population used in GS
is generally representative of the genetic diver-
sity used in breeding programs (wide range of
allelic diversity and genetic background). There-
fore potential applications at the scale of a whole
breeding program are expected to be more effi-
cient than any MAS approach derived from QTL
studies (Heffner et al. 2009; Heffner et al. 2010;
Jannink et al. 2010). GS approaches, which have
not yet been tested in sugarcane, appear to be
very attractive for breeders, given the high com-
plexity of the sugarcane genome and the complex
quantitative nature of the majority of agronomic
traits they need to tackle.
Conclusion
Sugarcane is one the most efficient plants for
biomass production. However in recent decades,
the increase in yield has been slower than in other
Search WWH ::
Custom Search